Joule Thief

[jeanna]

I made a new coil today.

It is made on a toroid and primary bifilar just like the 'widow'.
I have made many of these so far and I am confident of what to expect.

In the usual turns ratio short hand It is
5T,8T,48T

This usually nets about 40v, maybe 50 if I open up the transistor a lot. (and we know I do not like to do that.)

Because of the way I wound the 48 turns of the secondary, however, I got 90v from the 48T.
This is pretty good for this little toroid well, it is about double. 

Here is why... I think.
I wound it in a general way like the MK1, but the numbers of turns between turning points went up by a multiple of 8 which was the first number of turns.

So, I went up 8T
Turned and went down 8 again and continued up 8 more on the other side for a total of 16 turns,
turned and wrapped 3 turns in between each of the 8 on that last side for a total of 24 in that 3rd segment.

I am giving it the name
'MK1-jeanna Hybrid'

The idea behind this was to make a 3 cycle pulse.
I did not see any higher frequency... actually maybe lower??, but the voltage about doubled.
The amps draw was not more than 20mA, of course... this is the work of the transistor to make the pulses.

I am pleased because of the large voltage increase, but a little disappointed in the look of the wave which is very ordinary. (remember how I love those beautiful complex wave forms... oh well - good voltage.)
Xee2 showed that the wave form of the plain solenoid was great to light a fluoro tube, so I should forget about the pretty forms, I know.

Anyway, this is a lot of voltage for a few turns.
It may be the right idea finally.
I am about to make the other side of the cone into an hourglass, which is what I really want to be making, but I had this idea and needed to report, because the voltage is so high.

Maybe someone wants to try it?

<--------| 8T, --------|-------->16T, |<------------------------ 24T

This sign: | is the start and finish point of the secondary wires.

jeanna

[gadgetmall]

Very nice,

thank you ist . This is very heavy duty,
and
thank you Albert. (I like the looks of the cube too. I will check out where I can find a trophy case display cube. I don't think they do sports here but they do fish... maybe they do trophies for big fish.  )

jeanna

I live near East Carolina university and Granger Stadium ,Home of the Pirates . We have a Large Selection OF Lucite Cubes ranging  from Golf Ball to basketball cases ,even Football ones . there two piece .If there is an Interest i can get them and put up on Site if there not available in your area .


Al

[Yucca]

Wow this is a big thread!

Just a construction idea for exploring HF JT operation:

Did you know the BC547 or 2N2222 transistor is very fast and can easily go up to 100 MHz.

A ferrous core will limit frequencies to kHz region.

My idea is to wind a ~5cm diam drive coil using magwire, fairly thin, around 28AWG , say 20 turns.

Then use this as a core to wind a spiraling trigger coil around it, 20 wraps spiraling round the whole circumference of the drive coil. Can use 28AWG again.

If you have 26AWG or 32AWG then go with that.

Make the trigger coil flyleads be 180 degrees away from the drive coil flyleads.

Then using a fast transistor build a JT round it, mount the transistor and resistor (use carbon resistor) in the middle of the drive coil, just have the legs bent toward relevant flyleads.

The builder could then rightly claim to have "The Fastest JT in the WORLD!" 

Then it could only be beaten by using an RF transistor and Litz wire coils.

edit:also try swapping coil designation, so use spiraling wire as drive and coreloop wire as trigger. Also try a small electrolytic cap over power input lines, again mounted in coil centre.

[innovation_station]

HI YA BRO...

I THINK SLAYER007 HAS THE FASTEST!!



THE 2N2222A DOES 300MHZ.. 

I HAVE A FEW COILS THAT ARE VHF!!   NOT ON THE DRIVEING SIDE BUT ON THE OUTPUT SIDE...

YOU CAN HEAR THEM RAMP UP .....   UNTILL YOU NO LONGER CAN HEAR THEM .. THERE WAY OFF MY 20MHZ SCOPE

IST

THERE FAST LITTLE CRIDDERS ...

THE INPUT VOLTAGE AS I RECALL  EFFECTS THE SPEED ...  AS MY VOLTAGE DROPED IN MY SUPPLY I COULD GRADUALLY SEE THE WAVE ON MY SCOPE . 

I HAVE 500 MORE CORES SMALL ONES TO WIND AND FIND THE BEST WINDING CONFIGURATION .. 

THOS WILL BE TESTED FIRST .. TO SEE WICH CHARGES THE ULTRA CAPS THE BEST .. THEN I WILL WIND THAT COIL CONFIG ON THE LARGE CORE ...

PROBLEM IS I ALREADY HAVE A WINDING CONFIG IN MIND FOR THE LARGE CORES ..   AND BELEAVE ME THERE AINT NUTTIN HARD BOUT HOW THOSE ARE WOUND 

LARGEST FACTOR ALL WIRES ALL THE SAME LENGTH!!  AND TWISTED AS 1 WIRE ... ALL THE WAY AROUND ... 1 LAP 2 LAPS 3 LAPS .... DEPENDS UPON THE FREQ YOU DESIRE ON THE OUTPUT SIDE .. 

@ J   I WOULD LOVE TO WIND AND USE THE JEANNA AND MK DESIGN .. 

BUT I THINK FOR SOME REASON THE TURNS MAY NEED TO MATCH THE TURNS OF THE CORE IN MY CASE ...  5 TURNS AROUND MY TOROIDE ON INPUT IM THINKING CENTERTAPPED 5 TURN EACH WAY ...  ST WAS DOING THIS WITH SPECTULAR RESAULTS  SOME TIME AGO ... USEING THE DB107'S... 

[WilbyInebriated]

Wow this is a big thread!

Just a construction idea for exploring HF JT operation:

Did you know the BC547 or 2N2222 transistor is very fast and can easily go up to 100 MHz.

A ferrous core will limit frequencies to kHz region.

My idea is to wind a ~5cm diam drive coil using magwire, fairly thin, around 28AWG , say 20 turns.

Then use this as a core to wind a spiraling trigger coil around it, 20 wraps spiraling round the whole circumference of the drive coil. Can use 28AWG again.

If you have 26AWG or 32AWG then go with that.

Make the trigger coil flyleads be 180 degrees away from the drive coil flyleads.

Then using a fast transistor build a JT round it, mount the transistor and resistor (use carbon resistor) in the middle of the drive coil, just have the legs bent toward relevant flyleads.

The builder could then rightly claim to have "The Fastest JT in the WORLD!" 

Then it could only be beaten by using an RF transistor and Litz wire coils.

edit:also try swapping coil designation, so use spiraling wire as drive and coreloop wire as trigger. Also try a small electrolytic cap over power input lines, again mounted in coil centre.

i did something similar long ago, but i wound the coils directly on the transistor body. fast and small.